508 - Lowering Platelet Transfusion Threshold among Premature Infants: A Quality Improvement Initiative
Saturday, April 26, 2025
2:30pm – 4:45pm HST
Marwa M. Elgendy, University of Florida, Jacksonville, FL, United States; Gabriela Bianchi Leao, University of Florida College of Medicine, Jacksonville, FL, United States; Shereen Darwish, University of Florida College of Medicine, Jacksonville, FL, United States; Luisa Batine, University of Florida College of Medicine, Greenville, NC, United States; Agnes E.. Aysola, University of Florida, Jacksonville, FL, United States; Mark L. Hudak, University of Florida College of Medicine - Jacksonville, Jacksonville, FL, United States; Josef Cortez, UF Health Jacksonville, Jacksonville, FL, United States
University of Florida Jacksonville, Florida, United States
Background: Platelet transfusion at a higher platelet count threshold of ≥ 50,000/mm^3 is associated with a higher rate of death or major bleeding among preterm infants. Quality improvement methods are effective in implementing process changes to improve outcomes. Objective: Our primary goal was to reduce our threshold for most platelet transfusions to < 25,000/mm^3 based on our interpretation of recent evidence. We planned for a specific, measurable, attainable, relevant, and time-limited objective of decreasing the rate of platelet transfusion when platelet count was ≥ 25,000/mm^3 among preterm infants < 30 weeks’ gestation from a baseline of 60% in January 2020 to 25% in June 2024. Design/Methods: In our tertiary neonatal intensive care unit housed in an academic center that is also a safety net hospital at the University of Florida Health - Jacksonville, we identified key drivers and implemented quality improvement (QI) cycles. Key interventions included conducting a journal club to discuss the evidence that supported a lower threshold for platelet transfusion, providing education to NICU staff, modifying orders in the electronic medical record for platelet transfusions including precise indications, and activating a best practice advisory (BPA) warning when transfusing for platelets ≥25,000/mm^3 (Figure 1). We excluded infants with significant congenital anomalies. The proportions of platelet transfusions (number of platelet transfusions divided by total number of preterm infants) were plotted in run charts. As a balancing measure, we used the composite of death and/or severe intraventricular hemorrhage (IVH). We analyzed outcomes using statistical process control (SPC) chart (p-chart). Results: Among 400 infants < 30 weeks’ gestation, there were 163 platelet transfusions administered to 74 preterm infants. The proportion of platelet transfusions for platelet counts ≥ 25,000/mm^3 decreased from an aggregate baseline of 46.9% to 14.7% (Figure 2A). Similarly, the proportion of platelet transfusions for platelet counts ≥50,000/mm^3 decreased from 23% to 6.8% (Figure 2B). There was no difference in the composite outcome of death/severe IVH (Figure 3).
Conclusion(s): Our QI initiative successfully decreased the proportion of platelet transfusions for platelet counts ≥ 25,000/mm^3 without increasing the rate of death/severe IVH. By effectively lowering the threshold for platelet transfusion to < 25,000/mm^3, we also decreased the proportion of platelet transfusions for platelet counts ≥ 50,000/mm^3.
